Case Study of 8 COVID-19 Patients with Anti-SARS-CoV-2 Antibodies in CSF and Blood-Brain Barrier Disruption

A case series report published in Neurology: Neuroimmunology & Neuroinflammation describes eight COVID-19 patients with encephalopathy prolonged comatose or stuporous states (Alexopoulos et al., 2020). These patients were enrolled in the study from the ICU of the University Hospital in Athens, Greece and showed “signs of encephalopathy, including altered mental status, confusion, delirium, stupor, or coma, either before intubation or after sedation withdrawal when their respiratory status had improved.” There were 4 men and 4 women enrolled and they were between the ages of 57 and 84. None of them had a history of neurologic disease. None of the patients presented with neurologic symptoms at hospital admission except for one, who had agitated confusion. These patients had prolonged stays in the ICUs with an average stay of 36.7 days. Despite improvement of pulmonary function and extubation, “3 patients remained comatose and 3 stuporous, whereas 2 others still required sedation. Three patients eventually died.”

All patients had a high titer of anti-SARS-CoV-2 IgG in serum and all had at least some level of antibodies in the CSF, as confirmed by ELISA assay. Half the patients had a very high titer of anti-SARS-CoV-2 IgG in CSF, comparable to that of serum (detectable at 1:100 dilution). No anti-SARS-CoV-2 IgG was observed in 20 control CSF samples from patients with other CNS diseases. One patient had an IgG index of 1.85, which denotes intrathecal synthesis of IgG. Three other patients had an albumin index of >20×10^-3, which denotes blood-brain barrier (BBB) disruption. Four patients were positive for 14-3-3-protein in the CSF, which denotes possible onset of neurodegeneration vs inflammation. “All patients were negative for autoimmune encephalitis antibodies.”

In their discussion, the authors note that all the patients’ CSF was negative for SARS-CoV-2 by PCR, which is supported by other studies and rules out direct neuroinvasion as a pathophysiologic mechanism in this case report. Previously, I have written about an in vitro study done using iPSC-derived BrainSpheres that demonstrated SARS-CoV-2 can theoretically invade neurons. I have also written about post-mortem reports that have demonstrated that the virus does make its way into the brain, albeit much less so than other organs. It seems that despite these findings of neuroinvasion in vitro and post-mortem, a lot of the reports describing encephalitis or other severe neurological symptoms in COVID-19 patients are pointing more in the direction of indirect effects of COVID-19 causing neurological symptoms (via cytokines or multiorgan failure). Still, the potential for direct invasion cannot be completely ruled out. The virus is able to infect endothelial cells, an important component of the BBB, which may lead to disruption of the barrier. There is still a lot we do not know about how the virus mediates severe neurological symptoms, but each new case report provides further insight into this complicated (and fascinating!) issue.